The invention concerns a method of controlling an apparatus for supplying air pressure to a patient suffering from sleep problems.
The invention also concerns an apparatus for supplying air pressure to a patient suffering from sleep problems.
These sleep problems are respiratory and tend to waken the patient inopportunely.
They are for example apnoeas, hypopnoeas, acoustic vibrations or snores, or limitation of the respiratory flow, due to a narrowing of the upper airways of the patient.
The document U.S. Pat. No. 5,458,137 describes a method and a device for controlling respiration in the case of sleep problems, which use multiple and variable pressure levels.
A pressure source supplies a breathable gas compressed at a relatively low pressure to the airways of the user.
Pressure sensors monitor the pressures and convert them into electrical signals.
The electrical signals are filtered and processed in order to extract specific characteristics such as the duration and energy levels.
If these characteristics exceed chosen duration and energy level thresholds beyond a minimum time period, the microprocessor indicates the presence of a sleep respiratory problem.
If a chosen number of these events appears during a chosen time period, the microprocessor adjusts the pressure supplied by the source.
The document U.S. Pat. No. 5,490,502 describes a method and an apparatus for optimizing the controlled positive pressure in order to minimize the air flow coming from a generator while ensuring that flow limitation in the airways of the patient does not take place.
Provision is made therein to detect flow limitation by analysing a respiratory flow wave.
As soon as the presence of a flow limitation has been analysed, the system determines an action to be performed for adjusting the controlled positive pressure.
The pressure is increased, reduced or maintained depending on whether flow limitation has been detected and according to the previous actions implemented by the system.
The documents U.S. Pat. No. 5,335,654, EP-A-661 071 and EP-A-651 971 should also be cited.
The invention aims to improve the methods and devices of the state of the art, to automatically and continuously adapt the delivered pressure to the state of the patient and to anticipate and prevent the appearance of problems.
A first object of the invention is a method of controlling an apparatus for supplying air pressure to a patient suffering from sleep problems such as apnoeas.
A second object of the invention is an apparatus for supplying air pressure to a patient suffering from sleep problems such as apnoea, implementing the supply method.
The patient wears a mask by means of which air under pressure is supplied to his upper airways by the apparatus.
According to the invention, a control algorithm is provided using an output flow signal from the apparatus for detecting apnoea, hypopnoea, flow limitation events and leakages, and using the analysis of an item of pressure information for determining the presence of snoring, also referred to as acoustic vibrations.
The pressure supplied to the upper airways of the patient by the apparatus can be maintained constant, be increased or reduced according to the determination of the event which has been performed by the control algorithm.
Thus, if no respiration is detected by the control algorithm within a predetermined minimum time depending on a calculated mean respiration time, the presence of an apnoea is determined.
This predetermined minimum apnoea detection time is for example equal to a time constant, for example 10 seconds, added to a proportionality factor multiplied by the calculated mean respiration time, this factor being for example equal to ⅝.
For each apnoea, the output flow signal is amplified and filtered in order to determine the presence or absence of cardiac oscillations.
If cardiac oscillations were detected during the last elapsed time interval, for example equal to 5 seconds, then the apnoea is classified as being central and no control takes place in the algorithm.
If no cardiac oscillation was detected in this time interval, the apnoea is classified as being obstructive, and the pressure is increased by a predetermined value a first time and, during the same apnoea, twice more regularly, for example every 15 seconds.
The control algorithm compares peak-to-peak flow variations during the latest respiration of the patient with respect to a predetermined number of previous respirations, for example equal to 8.
After each respiration, a classification is performed into:
normal respiration, if the last peak-to-peak flow value is within a given range with respect to the mean value over the previous 8 respirations, for example from 40% to 150% or 140% thereof;
hypopnoeic respiration, if the last flow value is below this range;
hyperpnoeic respiration, if the last flow value is above this range.
A hypopnoea determination is made if hypopnoeic respiration detection takes place during at least a given time, for example 10 seconds, and terminates after a given number of normal or hyperpnoeic respirations, for example equal to 2.
A hypopnoea determination causes a given increase in pressure, for example 1 cm H2O first, and then, during the same hypopnoea, an increase in pressure by another given value, regularly, for example 0.5 cm H2O every two hypopnoeic respirations.
The control algorithm analyses and compares, respiration by respiration, the waveform of the respiratory flow with a sinusoidal waveform of the same period and same gradient.
After the comparison based on two flow form criteria, each respiration is first classified as normal, intermediate or limited flow.
A final classification, based on the combination of the flow classification and the occurrence of snores, changes the classification of respirations from normal into intermediate, respectively from intermediate into limited flow respiration.
Processing is decided upon when a certain number, for example 2, of successive limited flow respirations or a certain number, for example 5, of successive intermediate respirations take place after for example two normal respirations.
This processing causes a given increase in pressure, repeated regularly a certain number of times, for example 0.3 cm H2O three times every two respirations.
For each respiration, the pressure signal is amplified and filtered in order to detect the presence or absence of acoustic vibrations or snoring.
A determination of a valid snore is made by the control algorithm if the detected acoustic vibration occurred at least for a certain time, for example 7% of the mean duration of the last three respirations, and with a period less than a factor proportional to this mean time, for example 120% thereof.
In the case of a valid snore, the algorithm increases the pressure by a given value, for example 1 cm H2O, if the last control due to a snore took place more than a given time previously, for example 1 minute.
A mean leakage is determined as being equal to the mean flow during respiration.
The control algorithm continuously compares the current leakage with a leakage limit, it being possible to regulate said limit from the pressure.
If the current leakage exceeds the limit, all pressure increase controls generated following event detections are disabled.
After detection of an apnoea or a snoring event or a hypopnoea control or a processing decision, the algorithm will reduce the pressure by a given value, for example 0.5 cm H2O, in a first step after a given time, for example 5 minutes, and regularly for the following reductions, for example every minute.
A given maintenance pressure, for example 8 cm H2O, is supplied by the apparatus if no respiration has been detected during a given time, for example two minutes, or if the pressure supplied has been greater than or equal to a given value for a given time, for example 17 cm H2O for 10 or 30 minutes.
One advantage of the method is an automatic adaptation of the detection criteria to the respiratory characteristics of the patient.
Thus, any modification of the respiratory rhythm is taken into account by the algorithm for performing the detection.
The fact of involving a mean value of respiratory cycle time over a certain number of previous respiratory cycles has the effect of variations in the cycle and respiratory amplitude being tracked regularly and better detection.
The invention will be better understood from a reading of the following description, given with reference to the figures.